Hello Gooddata team, We want to know what is the ...
# gooddata-cn
s
Hello Gooddata team, We want to know what is the difference of the errors in case of two different scenarios we query the data using attributes. This is what we see in logs and want to know the exact scenario these errors will be thrown to handle them accordingly. We have vague idea on when the AntiChain errors comes into picture where we query with unrelated attributes but
Genera error
is not a clear message to know what exactly is the error here. Can you please elaborate? Scenario-1: error message:
A result cache error has occurred during the calculation of the result
error detail:
ProcessError(errorType=CLIENT_EXCEPTION, message=Antichain='[attribute/Account_StartDate.day/workspace_id, attribute/FinancialInstrumentTypeCode/workspace_id, attribute/GLActivity_AsOfDate.year/workspace_id, attribute/Transaction.AccountKey/workspace_id]' has no witness)
Scenario-2: error message:
A result cache error has occurred during the calculation of the result
error detail:
ProcessError(errorType=CLIENT_EXCEPTION, message=General error)
i
Hi Sunil Kumar, Your assumption is quite correct. You can find more details in this post. Iā€™d suggest to use the validObjects API here as well. The second error is quite generic, can you share more details with us, please? What exactly let do this issue, please? Were you simply trying to render existing Insight, or were you adding a new attribute to your analytics?
s
Sorry for the late response, I missed your reply. I was trying to analyze with new attributes and measure combinations and got this error. I just want to know in what scenarios this error is thrown or else this error is a catch all and will be thrown for any sort of unhandled errors?
j
this is the error message for my recent query
the query is to fetch the following attributes: [ { "id": "AccountId" }, { "id": "FinancialInstrumentRealEstate.TenantClientOrgId" }, { "id": "FinancialInstrumentId" }, { "id": "Account.SourceAccountId" } ]
s
Can you share the LDM model for the attributes you selected @James Lee? that way Ivana can explain in what usecase this can happen
I think in the above case there is a witness available
j
theres no ldm model, the above are attributes and we pass no filter models to the query since we're testing the combination of attributes only
sdk.tables.for_items( workspace_id=workspace_id, items=[Attribute(local_id=name, label=name), ...], filters=[], )
something like that where we pass in 4 attributes to the items parameter for each of the 4 attribute ids mentioned above
m
Have you tried to find the trace id / result id in logs? Maybe there will be other log with more details.
r
Hi all, have you had a chance to look into Martin's recommendation above? šŸ™‚
s
@Martin Burian sorry for the late response.. we did check it that time and that says General Error too is what I remember
r
Hi Sunil, essentially yeah, the "general error" is the high level error that gets reported, however the reason Martin is asking is that usually, even these catch-all errors are only a result of a different component reporting a more specific error, coupled with an error code that gets interpreted into a "general error". So often enough, digging down via trace ID should lead you to a more specific (or well, sometimes at least longer) error message to go by šŸ™‚